The 10 days of Blogging past: Day 9

In the penultimate post for the 10 days of Blogging past is one of my first posts, yet one of the more influential in my blogging, "Concurrent:Active Request Limit (or how I finally learned to know what my system was saying)" from 8/16/12 with 126 views. 

While the title was certainly a mouthful, and a parody of Dr. Strangelove, more importantly it signaled the start of me looking deeper into "normal" things going on that I had been taking for granted.  Just because I've spent six or seven years seeing it happen didn't make it "normal" or "okay"!  What other type of articles came out from this train of thought?  Well, I would say that investigating RMAN when it was hanging up on us and also trying to figure out why Fixed Assets would error infrequently are top "normal" things that I started digging a bit deeper into.

So when I finally researched a bit why a request was pending in the Conflict Resolution Manager with an error message barking about 'Concurrent:Active Request Limit' being set, but I knew full well it couldn't be set, I was surprised to find out it actually WAS set.  Again, something I thought I knew, but the troubleshooting and research had to be done in order to make sure the set of facts I knew were in fact......facts.  Okay, maybe that's too many facts to get straight, but you get the picture.

On the blogging front, I think this has received so many hits because of how much text from the error message I put into the post which makes it very searchable if other people are trying to investigate their own issues with this.  The routine way this article gets hits makes me think this profile might be set as part of a patch or upgrade, or maybe legacy handed down from administrators way back in the day that might not make sense to the administrators of today.

The 10 days of Blogging past: Day 8

As we wind down the 10 days of Blogging past, we get to review a rather enigmatic post in "duration not active: A Tale of Two Articles" from 1/3/13 with 91 views.

This post captures two very different My Oracle Support articles, which may not strike you as a surprise if you've been on MOS for any period of time, but it strikes me as interesting that this is in my top THREE posts all time.  Why is that?  I think this is another reflection on what I've been trying to do for the last year, and that is present relevant and NEW content on a regular basis!  Sure, but still why would THIS get a lot of hits?  Well, this was a new error that popped up after we went live with a new Exadata BP so the assumption is this is an unintended consequence that many different people have seen after getting this BP in their system as well.  We never saw any negative results or impacts upon our system due to this message, so I have to believe this is not really a cause for concern but many other people have seen it and likely searched the web for experiences that others have had with it.

Additionally, this gives us a good insight into KB articles and how they can be 100% accurate but not really give any content or help to your end user.  Don't let your KB articles at work get like this, or your blog posts, or your tweets, etc.  Sometimes the best examples to learn from, are those that aren't the best examples in the first place.

The 10 days of Blogging past: Day 7

As we start hitting the home stretch of the 10 days of Blogging past, we will review a rather straightforward follow up post in "Learning more about DBA_ALERT_HISTORY" from 2/14/13 with 86 views.

As is often the case I learn the most about our systems when something "bad" happens or things go "wrong" in ways we had not anticipated.  I don't know if that is your typical experience in supporting an application, but that is certainly my experience!  While going to Exadata we were presented with the opportunity to uplift our entire infrastructure footprint, and one way we did that was going to a multi-database RAC node instance as opposed to a single database node instance that we were previously running.  We (and our DBAs) knew the flaws with the old design, but we had no idea all of the possible challenges that would face us with going to a RAC instance, of which this is obviously one of them.

As a result, I was able to learn more about what the table DBA_ALERT_HISTORY would (and wouldn't) log and the timing behind it.  It seems obvious that a node which is offline or disconnected for some reason wouldn't be able to log to a table if it can't be connected to it, but what was not obvious is that the remaining node would reconfigure as if the other node is not there AND not log something to the table.  Sure, the server comes back online/available it logs to the table, yet what happens during that time if you think DBA_ALERT_HISTORY will tell you the facts in a real time manner?  You'll get alerted way too late and oddly enough, the original message was logged to the table AFTER a new message was entered saying everything was fine.

This is a great point that dovetails with a lot of things other Oracle experts go through with having a hypothesis or knowing something might be true but being able to test it out and actually verify if that is a true statement/hypothesis or not.

The 10 days of Blogging past: Day 6

Returning to our 10 days of Blogging past review, today the post we're visiting is "Concurrent Manager holding lock on a report" from 5/23/13 with 82 views.

Despite the obvious lazy sounding title, but today I don't know what else I would've picked to make it sound better, this is the first post in the top five that we're reviewing so it has found an audience out there regardless!  Like a previous review, when I read this today it sounds very clear to me since I experienced it yet I know some of my inexperience is showing through when I'm trying to explain what is going on because it sounds muddled because I need to get better at mixing "how to" with a narrative style.

That said, I have shared many posts on the concurrent manager:

http://theoracleemt.blogspot.com/2013/05/icm-will-mark-this-request-as-completed.html
http://theoracleemt.blogspot.com/2013/08/concurrent-reports-not-completing-for.html
http://theoracleemt.blogspot.com/2013/08/concurrent-manager-bottleneck.html
http://theoracleemt.blogspot.com/2013/09/what-happens-when-your-concurrent.html

and several of them have enjoyed pictures to go with the actual post content, which were my first attempts at using images on my blog to enhance the narrative.  Great idea if you have your own blog!

On the more application side of things, this post (and some the investigation done) has come in handy many times when trying to understand problems with the concurrent manager and the processes they run.  This may even lead into a "part 2" type of post when I answer some of the lingering questions I have about the relatively simple technology that is the concurrent manager platform and I'll try to focus on finding out more information about it while solving its riddles.

Thursday Twitter - 12/26/13

Taking a break from the 10 days of Blogging past feature to bring you an armload of Twitter presents!  I have a lot more tweets to share with you, but it looks like that will happen next year.  :}

Announcements:

https://twitter.com/virtualbox/status/413414240058224640 - Oracle VM VirtualBox 4.3.6 is now available!

https://twitter.com/myoraclesupport/status/413509006746210304 - Become a part of the Accreditation Program for My Oracle Support!!

Training/Free Stuff:

https://twitter.com/oraclemagazine/status/413375622225088513 - The newest edition of the Oracle magazine is now available, and you can sign up for it too!

https://twitter.com/enkitec/status/413379697334968321 - Karen Morton is having a SQL tuning class soon!!

https://twitter.com/oracleblogs/status/413661820915027968 - Oracle SOA Black Belt cheat sheets available for download!

https://twitter.com/kylehhailey/status/413699469717155840 - Kyle Hailey gives us his UKOUG presentation!!

https://twitter.com/oracleblogs/status/413707743707811840 - SQL Analytics Part 2, which is interesting enough you should review part 1 as well!

https://twitter.com/oracleblogs/status/413724437486534656 - A $5 e-book bargain sale!!

https://twitter.com/enkitec/status/413755922645671936 - Carlos Sierra is offering training on the SQLTXPLAIN tool.

https://twitter.com/oracledbsupport/status/413774510132056064 - Oracle SuperCluster webinar.

https://twitter.com/oracle_edu/status/413783212440453120 - Oracle University reminds about the trainings they have available.

https://twitter.com/oracleblogs/status/413858083358932992 - Virtual Sysadmin Day is happening again!!

https://twitter.com/oracleblogs/status/413963485262135296 - Oracle Solaris Hands-On Lab is now available!

https://twitter.com/oraclepress/status/414072989794435073 - Learn the difference between SQL and NoSQL in this free chapter download!!

https://twitter.com/enkitec/status/414087457689202689 - Do you want 20% off your training from Enkitec?

General News/Information:

https://twitter.com/martinberx/status/413359352167161856 - Martin Berg comments on a tweet, which brings out some good information about PGA limits.

https://twitter.com/oaug1/status/413378853743374336 - OAUG asks what your big accomplishments were for the year.

https://twitter.com/oracle/status/413387158309724161 - Oracle Excellence Award winners!!

https://twitter.com/yvelik/status/413399207819501568 - Yury Velikanov was interviewed by IOUG!

https://twitter.com/kylehhailey/status/413426722739351554 - Oracle beats Q2 estimates!

https://twitter.com/oraclehardware/status/413624773118550016 - Oracle Migration Factory!

https://twitter.com/leight0nn/status/413708214010929152 - Leighton Nelson asks a question about redo log threads, and starts a good conversation.

https://twitter.com/oracleexalogic/status/413715374077902850 - I've shared that the new Exalogic server has been released, but 720 cores per rack?!?

https://twitter.com/infosecuritymag/status/413781605950312449 - Who hasn't heard about the incident with Target?  This article makes it sound like it was an inside job.

https://twitter.com/oracleblogs/status/413812675127566336 - Installing Oracle Ops Center.

https://twitter.com/arupnanda/status/413898079784669184 - Arup Nanda became a published author 11 years ago!!

https://twitter.com/oracleblogs/status/413902779028361216 -Introduction to SOA!

https://twitter.com/oraclecloudzone/status/413930527700553728 - The cloud conversation is changing.

https://twitter.com/oracle_ebs/status/413931786386038784 - Get help when Procurement performance is bad even after Purge Obsolete is run.

https://twitter.com/chrisantognini/status/413993990984564736 - See what parameter Christian Antognini thinks should be set as default in 11.2.0.4.

https://twitter.com/hdeiby/status/414047794115014658 - Latin America, Oracle and Exadata!

https://twitter.com/oracle/status/414059552212910080 - Oracle's Technology Year in Review!!

https://twitter.com/leight0nn/status/414063377632096257 - Leighton has a good insight about working with MOS!

https://twitter.com/thatjeffsmith/status/414073992426053633 -What is your request for SQL Developer training videos?

https://twitter.com/uwehesse/status/414081777222049792 - Uwe Hesse got some feedback from a class which is pretty awesome!

https://twitter.com/oracledatabase/status/414093757567758336 - A podcast on SQL Developer 4.0 featuring Jeff Smith!

https://twitter.com/kevinclosson/status/414097632106016768 - Kevin notices that Exadata's Flash Cache Compression needs a software license.

https://twitter.com/chrisantognini/status/414101153014890496 - Christian tells us that there is a new 12c feature with PL/SQL invoker's rights.

https://twitter.com/acolvin/status/414135945987108864 - Andy Colvin shows us a very confused Exadata V2!

https://twitter.com/oraclebase/status/414144703437996032 - Tim Hall has a very Christmas-y picture for us.

Conference News:

https://twitter.com/inoug_org/status/413754074803081216 - INOUG has a kick off meeting!

https://twitter.com/thatjeffsmith/status/413763961331662848 - Jeff Smith shares the top 4 KScope sessions he wants us to go to this year.

https://twitter.com/richardfoote/status/413779281450258434 - Richard Foote will be presenting at ODTUG's KScope as well!!

https://twitter.com/hotsosgg/status/414062842812182528 - Register for the Hotsos Symposium this year and see ME!!

Social Media/DevOps:

https://twitter.com/clicktips/status/413710489764368384 - Melinda Samson shares some tips about Google Analytics and benchmarking.

https://twitter.com/clicktips/status/414045737056354304 - Get to page #1 on Google without using/thinking about SEO.

https://twitter.com/clicktips/status/414086367417602048 - I didn't know you could segment your blogging traffic this way!

https://twitter.com/dbakevlar/status/414090449431715840 - Kellyn reminds us about getting an about.me page.

The 10 days of Blogging past: Day 5

Today for the Christmas episode of 10 days of Blogging past, I will be reviewing the "Supplier issue in Expense Report Import" post from 4/23/13 with 81 views.

I can't say that I'm really surprised this is in my top ten, as this has come up several times at work and highlights not really a design flaw but reflects a different way of thinking about problems that would come up eventually.  The good news was that there were many different articles on My Oracle Support which were able to direct me to a resolution rather quickly, but this is a sign that the issue has come up for a LOT of people over the past few years; that reveals what I think of as the different way of thinking that makes the entire process bomb out instead of allowing for issues and continuing to process what is not in error.  We've had processes similar to this designed over the past few years, and because of our experiences with programs like this we choose either to allow it to process with an error on the output, or a secondary program in the report set is built to kick out the errors so it is clearly visible to the business that there was an error with the processing.

We can also see what is a secondary difference in thinking by requiring additional "IT" work to provide the business the exact root cause when it is fairly easy to identify with the scripts provided in MOS and in this post.  If it is possible to clearly tell what has gone wrong, and how it might be able to be fixed, we should make every effort to do so in order to provide better service to the business and reduce unnecessary work for IT.

The 10 days of Blogging past: Day 4

The 10 days of Blogging past continues today with our post "Your Oracle DB not playing nice with Hyperion products" from 4/27/13 with 72 views. 

This is a rather straightforward post, hiding a deceptively complex issue.  How many of you know that you have alerts on 100% of your disk volumes?  Is /flash monitored on 100% of your systems?  Do you know which disk volumes are critical to the continued success of your application?  If you're responsible for Hyperion products, are they linked to an Oracle database that you're monitoring appropriately?

Going off on another tangent, it's clear from this post that "something" was going on and it was this "something" which caused us to do a wholesale reboot of our environment.  Now part of the process needs to be a way to investigate further/faster next time so we understand more about how our environment is working together to communicate and provide services to our end users.  As this never happened again, and we upgraded our Hyperion environment as well, I doubt I'll ever get a chance to replicate this (or see it replicated) but it does make one wonder if this "flaw" exists now so I have an outstanding item to test next year.  This was also a type of sequel from another post, where one server worked and another didn't in a RAC setting which didn't make a whole lot of sense either, but having this happen to us months earlier helped us to make better troubleshooting choices to get to our problem resolution a lot quicker next time.

The 10 days of Blogging past: Day 3

Continuing the 10 days of Blogging past feature, the post we will be reviewing today is "System Wait Events Part 2" from 10/17/12 with 71 views.  This is a bit of nostalgia for me, as this was one of my first real "serious" posts after we installed the Exadata platform in our environment and it has actually lead to some discussions with individuals from Oracle about the problem we encountered.  I can say that about no other blog post I've written, and while I fancy that maybe if I can share my thoughts and problems with my ever growing network of Oracle employees I can have some type of influence, maybe the sharing of this event will have that type of influence and help Oracle improve their design or systems in some way.  That's an incredible thought!

On another train of thought, looking at this post makes me think about better ways to format text with the Blogger tool.  Also, I think I've grown at explaining problems and giving attribution a whole lot better because this seems a bit rough around the edges plus I completely missed giving credit for this the first time around due to my inexperience.  This leads into another good tip: you need to start writing to find your voice, understand what your style is, allow your abilities to grow with time because none of that will happen if you don't start writing and sharing your passion.

On the technology front, when something like this happens it can rekindle your passion for learning and I know this had a similar effect for me as I started diving more into the ZFS, cells, and Exadata in general after this happened!

The 10 days of Blogging past: Day 2

Continuing on the new feature introduced yesterday, the second post we're going to review for the 10 days of Blogging past is "Disconnected Concurrent Manager" from 9/29/13 with 61 views.  Right off the bat, I notice that this post has been online for almost 8 months less than yesterday's but it has more views!  Why is that?  I believe it is absolutely due to the amount of error messages in the post and how many times the messages pop up for people in their daily lives which leads to them searching online for a fix.  Sure I had a clear example in the post from day one, but it is targeted at just the Fixed Assets module and this post has errors from a lot of different areas/modules so it has a wider appeal.  That's a lesson learned for me just now as I compare the posts!

So what if you aren't blogging, what can you get out of this review?  Well, great question!  Breaking this down a bit, I can see that the log and output files have some very clear indicators of a problem and that these indicators will probably crop up time and time again when you have similar problems.  Why is that a big deal?  Do you have any tools in your environment that can look at the log files?  If you do, can you fine-tune them to be looking for some of these errors and then alert you when they happen?

Fast forward a few months after this event, and the next time we had work planned for our network guess what?  We were a lot more conscious of how network outages/interruptions affected us, so we went the extra mile to protect ourselves from another such incident.  This is a take away that anybody can enjoy!

The 10 days of Blogging past: Day 1

This holiday season I'm going to take time to review the top 10 blog posts, as measured by hits per post, and offer new information in some way so you get something out of it too!

The first post we're going to review is "Purging concurrent and Fixed Assets" from 2/1/13 with 60 views.

While I had already created the alert before putting up the blog post, I learned something afterwards by having our team act upon the alert: it is possible to CREATE a problem instead of resolve one taking action.  The alert was created to look at problems when the Calculate Gains and Losses or Depreciation Run concurrent reports error in the Concurrent Manager within the past 2 hours, but it does NOT require action to be taken in our example since we don't run into problems until we get close to the request getting purged within 3 DAYS time.  I've seen a few times when we clear out fa_book_controls.mass_request_id too quickly it seems to cause some issues.

The lesson to take away from this is, building alerts is great as long as they are followed with clear documentation and make sense 100% of the time.  If an alert goes off too much, or gives a false positive, you are setting yourself up for failure eventually.

EBS 12.2.3 is released!!

Okay, I'm cheating a little since I shared a tweet about this yesterday, but being an "apps guy" I think this some pretty big news!  Follow this link to Steven Chan's EBS blog where he has the files available, the readme linked, says what is included like a new look and feel plus a Worklist feature which sounds pretty darn cool to me!!  On top of all of this, are the usual upgrade path documents that you should be reviewing and getting familiar with if you're not on R12 already (or if you have been on a lower version for years, and want to be getting the latest and greatest version of R12).

Thursday Twitter - 12/19/13

Getting ready for the holidays, I'm starting the gift giving early with these tweets!

Announcements:

https://twitter.com/JLOracle/status/411837449820585984 - Wikipedia is asking for donations, and Jonathan Lewis asks us to think of giving to them.  Who out there doesn't use this site, even for looking up the most trivial things?  Not a bad idea, not bad at all.

https://twitter.com/OracleBlogs/status/412665332550361089 - EBS 12.2.3 has been released!!

https://twitter.com/OracleCloudZone/status/412958425195499520 - Oracle Exadata Elastic Cloud X4-2 has been released too!!

Training/Free Stuff:

https://twitter.com/brost/status/412309649983942656 - Bjoern Rost asks a simple question about what people think the 3 more important things to learn about RAC, and starts a pretty good conversation!

https://twitter.com/OraclePress/status/412623407742980096 - Download a free chapter of the Applications Development & Extensibility Handbook from Oracle Press!

https://twitter.com/OracleDatabase/status/412633239141232640 - Oracle WebLogic and Cloud 12c Integration webinar.

https://twitter.com/Enkitec/status/412655775706652672 - Learn how to patch Exadata from the experts at Enkitec!

https://twitter.com/IOUG/status/412718833242484736 - A very high level break down of 12c basics courtesy of IOUG and Enkitec.

https://twitter.com/DBAKevlar/status/412770111221948416 - Kellyn comes through with 29 free books on several hot topics right now!

https://twitter.com/OTNArchBeat/status/412940274881802240 - This post on ADF mistakes is a. really good and b. has links to other conference papers/articles presented at UKOUG Tech this year!

https://twitter.com/ORCLGoldenGate/status/412987323140550656 - A new interactive e-book has been released to showcase Oracle Data Integrator 12c and Oracle GoldenGate 12c entitled "The Path to the Future".

https://twitter.com/fritshoogland/status/413019131621814272 - Part 2 of "When the Oracle wait interface isn't enough" by Frits Hoogland.

https://twitter.com/Oracle_Edu/status/413044880420524033 - Oracle University makes a good observation while including a Pinterest link?

https://twitter.com/OracleCert/status/413074326737276928 - Oracle Certification wants to help you gift this holiday season!

https://twitter.com/DBAKevlar/status/413094643027279872 - Join RMOUG!

https://twitter.com/Enkitec/status/413314367875067904 - Learn more about GoldenGate memory usage.

https://twitter.com/OracleBlogs/status/413362562861133824 - A teaser for the new ebook on Securing Oracle 12c.

General News/Information:

https://twitter.com/oraclehardware/status/412145026240888832 - Oracle and optical links?

https://twitter.com/History_Pics/status/412315576565235713 - Netscape Navigator 1.0 turned 19 this week.  Wow.

https://twitter.com/Oracle_EBS/status/412478466605735936 - Are you on 11.5.10?  Here are the Extended Support required patches!

https://twitter.com/OracleBlogs/status/412590181217484800 - Do you know why OPTIMIZER_FEATURES_ENABLE should not be set?  Find out!

https://twitter.com/DBAKevlar/status/412600012120793088 - Kellyn Pot'Vin wins the tweet of the week award.  :}

https://twitter.com/BrianPardy/status/412630143568728064 - Brian Pardy is seeing some odd behavior from 12c.  Have you seen this too?

https://twitter.com/grantunez/status/412838659516162048 - How good does your profession look for the future?  Rene Antunez shares some projected growth figures from the US Department of Labor.

https://twitter.com/karlarao/status/412935409492979712 - Karl Arao shares a way to easily forecast Exadata storage space.

https://twitter.com/DBAKevlar/status/413015970966302720 - Learn how Oracle offers some built-in savings with the database.

https://twitter.com/OracleMiddle/status/413096300633411584 - Want to learn more about double blind identity?  Read the latest blog from Oracle Middleware.

https://twitter.com/OracleBlogs/status/413104084686995457 - The most used Payables Knowledge Content this year!

https://twitter.com/DBAKevlar/status/413150739708407808 - Kellyn was interviewed in the Oracle Magazine!

https://twitter.com/grantunez/status/413165917543997440 - Tips for how to change the language in Grid Control.

https://twitter.com/brost/status/413263470759772160 - I wonder.  Do you think Fujitsu ordered enough pizza from this place?

https://twitter.com/OracleCloudZone/status/413310257063272448 - Rex Wang breaks cloud myths!

https://twitter.com/OracleBlogs/status/413331881082773504 - A new Oracle ACE!

https://twitter.com/kylehhailey/status/413353208979787776 - Kyle Hailey has an insight which expands on something I stumbled upon earlier this year, and brought forth in my presentation in Pittsburgh!

Conference News:

https://twitter.com/Enkitec/status/412708626780803072 - Don't forget to register for Enkitec Extreme Exadata Expo 2014!  Call for papers is still open too!

https://twitter.com/markrittman/status/412910273667039232 - The Rittman Mead BI Forum has opened their call for papers too!

Social Media/DevOps:

https://twitter.com/ClickTips/status/411853783925133312 - What are Twitter's 7 deadly sins, and how do you avoid repeating them?

https://twitter.com/potus98/status/412077379037573122 - Mainframers interested in DevOps?  That's awesome!  Everybody should be interested!

https://twitter.com/ClickTips/status/412594044045258752 - Using Google Analytics Trackbacks!

https://twitter.com/ClickTips/status/413031600658460672 - Additional tips for generating content?  Yes please!

https://twitter.com/dba_hba/status/413236225940520960 - Allan Robertson shares a site trying to build the learning base for SysAdmin.  He also shares another link that has a scorecard with 32 topics where a SysAdmin/group should excel.  How well do you do?  https://twitter.com/dba_hba/status/413242744853626880

Following up with Karen Morton

I've posted several times about the webinar's that Karen Morton hosts, but I've never provided you the follow up posts that she adds to get the videos or presentations that she used!  There's no time like the present, and actually, let me call this my present to you.  I've attended most of these, and the ones I've missed are going to be my homework over the holiday break.  :}

http://karenmorton.blogspot.com/2013/05/follow-up-using-optimizer-hints-for.html

http://karenmorton.blogspot.com/2013/11/webinar-follow-up-everyday-oracle-pro.html

http://karenmorton.blogspot.com/2013/09/webinar-follow-up-execution-plans-learn.html

http://karenmorton.blogspot.com/2013/07/my-day-with-ohio-oracle-users-group.html

http://karenmorton.blogspot.com/2013/07/follow-up-visual-sql-tuning-webinar.html

OPT_PARAM in-line hinting

As I learn more about writing SQL queries and system parameters, every so often I'll come back to some old blog posts from other authors.  This type of review allows me to get a handle on something that I didn't understand at first, and then share it with you after that.  Tim Hall created this post about a badly performing query throwing ORA-00600 errors, and the suggestion was to alter a system parameter for the session, but instead he chose to actually use the OPT_PARAM hint within the SQL query.  I thought it was smart enough to save for later and I suggest you read the article, and the other post that he linked within as well.

Date Formats and Oracle Alerts

I've been on an Oracle Alert kick lately with a few posts on the subject, and here comes another article with an different error message.  Looking for reports in Error leads me to find another Oracle Alert, but I can't see exactly what the issue is off hand because choosing the Verify option yields no errors.  When I choose Run, it is a different story as the alert gives me the following error message:

APP-ALR-04108: SQL error ORA-01850: hour must be between 0 and 23

Interesting to say the least.  What did I find?  Investigating further I found that the select statement for the alert is pulling in data from a table that has non-numeric characters and trying to apply a date format function to them which is making it so the alert will never run right.